Edit: Pretty much all conjectures in my question have been proved to be incorrect. I started fresh with a new approach. New question is here.

Original question:

I'm having an issue with a touchscreen LCD assembly.

The touchscreen normally works fine, but if I mount it in a housing with a gasket the taps sometimes register in the wrong location.

The incorrect taps only seem to occur when I'm tapping the screen. They don't occur on their own.

This is the screen I'm using: NHD-4.3-480272MB-ASXV-CTP (note the thumbnail is inaccurate)

I've tried different gasket materials such as neoprene rubber, and foam but the issue occurs with all of them.

lcd, installed front Inside view (without LCD) lcd gasket from inside

I placed the gaskets so that there is an ~1mm gap between the LCD active area and the gasket on all sides. The datasheets for this part don't have a recommendation for the gasket material or geometry.

Can anyone suggest a solution for this problem, or design guidelines I might use to come up with a solution?

The housing the screen will be mounted will be injection molded, so it's difficult to modify at this point. It doesn't really have any features besides an aperture, and screw holes for the bracket.

EDIT: New discovery, it doesn't seem to be the gasket after all. I realized that when the touchscreen is not installed, I'm holding it by the bracket and thereby grounding myself. If I set the touchscreen on the table and touch it with only 1 finger, the same problems occur. So, new question why does the touchscreen only work when I'm grounded? The RH is 65% in my office, so I don't think it's static.

EDIT2: I thought the power supply had negative connected to earth, but it doesn't. So I manually connected it, and that fixed the problem. ..But I still don't know why. Why does the touchscreen not work correctly when the supply is floating?

  • \$\begingroup\$ Have you double checked your wiring? Do you have another panel to replicate the problem on? \$\endgroup\$
    – user199402
    Jun 25, 2019 at 21:58
  • \$\begingroup\$ Yes I have 2 panels, same problem on both. The touchpanel has a built in controller and communicates with SPI iirc, so I'm not sure what wiring to check. \$\endgroup\$
    – Drew
    Jun 25, 2019 at 22:03
  • \$\begingroup\$ In the datasheet, it shows Pin 1 as VDD, and Pin 2 as GND for a 6-pin connector. Then it shows grounds on Pins 3, 29, and 36 on a 40-pin connector. Just good to start with making sure everything is connected properly. \$\endgroup\$
    – user199402
    Jun 25, 2019 at 22:11

1 Answer 1


Capacitive touchscreens with glass need a reference ground and same reference grounded film capacitor to work properly at the same potential. CM noise can be an issue when present but worse when it is a differential noise.

  • \$\begingroup\$ Interesting point, but battery powered devices have touchscreens with no issues so I don't see how this shouldn't be possible to fix without grounding the supply? \$\endgroup\$
    – Drew
    Jun 26, 2019 at 5:17
  • \$\begingroup\$ Your supply must not be EMI proof then Ground means some low impedance DC ref. not Earth Ground , but it may in your test. \$\endgroup\$ Jun 26, 2019 at 6:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.